1. Field of the Invention
The present invention relates to an image processing apparatus and method and, for example, to an image processing apparatus and method for correcting the gradation characteristics of a color image.
2. Related Background Art
A digital color copying machine has mechanisms for correcting the influence of ambient variations and variations caused by aging in a printer unit. As one of such mechanism, an automatic gradation correction function that uses an image scanner unit as a measurement device is known. The automatic gradation correction is performed by the following procedure.
(1) The printer unit prints out a test pattern having predetermined values of C, M, Y, and K colors. PA1 (2) The image scanner unit reads the test pattern. PA1 (3) The gradation characteristics of the printer unit are obtained based on the reading result of the test pattern, and a coefficient or table for correcting the obtained characteristics is calculated. PA1 (4) The calculated coefficient or table is set in a printer gradation correction circuit.
With the above-mentioned automatic gradation correction, the gradation characteristics of the printer unit can be stabilized.
However, the above-mentioned technique suffers the following problem. That is, when a CMYK test pattern is read upon execution of the automatic gradation correction, errors caused by variations in color filters of a CCD may be generated. This problem will be explained in detail below.
FIG. 2A shows the spectral sensitivity characteristics of the R, G, and B channels of a so-called 3-line sensor. To obtain an image with good color reproducibility by faithfully reading a color image, the spectral sensitivity characteristics of the R, G, and B channels have overlapping portions, as shown in FIG. 2A, so as not to form non-detectable wavelength ranges. Therefore, signal processing must be performed under the assumption that the spectral sensitivity characteristics of the R, G, and B channels overlap each other to some extent.
FIG. 2B shows the spectral sensitivity characteristics of the B channel and the spectral reflection characteristics of the Y test pattern. Since the Y test pattern exhibits a high reflectance with respect to the wavelength ranges of the R and G channels, the density of the Y test pattern cannot be detected using the R or G channel, and hence, the density is detected using the B channel. However, when the spectral sensitivity characteristics of the B channel, i.e,. the spectral transmission characteristics of the B color filter vary, as indicated by a broken curve, the density detection precision of the Y test pattern lowers due to the influence of the variation, and as a result, appropriate gradation correction is disturbed.
In another technique, N sets of known input signals An (print signals) are supplied to an image forming apparatus to form N sets of specific color patterns, luminance levels Bn of the formed patterns are read, and N sets of printer output signal values Cn are obtained based on the luminance levels Bn by, e.g., masking calculations. By adjusting image forming parameters such as masking parameters so that the N sets of input signals An roughly equal the output signals Cn, the stability of image quality is improved.
However, the above-mentioned technique suffers the following problem. That is, when image formation is repeated, toner turns into a powder and the toner powder becomes attached to carriers of a developing agent, resulting in a decrease in maximum density of a copied image. The decrease in maximum density narrows the color reproduction range. In this state, even when the image forming parameters such as masking coefficients are adjusted, a good image cannot be obtained. The deterioration of image quality due to deteriorated durability of the developing agent can be eliminated by exchanging the developing agent. However, when the developing agent is periodically exchanged in consideration of the deteriorated durability, the copy cost rises undesirably.
Also, in another technique, a specific pattern is formed on a recording medium in an image forming apparatus, the formed pattern is read to feed back the read data to the image forming condition such as .gamma. correction, thereby improving the stability of the image quality.
However, the above-mentioned technique suffers the following problem. In a device for reading an image, when the characteristics of color separation filters in a CCD deviate from ideal spectral characteristics, different image signals are obtained even when a pattern formed on a single recording medium is read. Therefore, even when the image forming condition is calculated based on such image data, an optimal image output cannot often be obtained from the image forming apparatus.
As is conventionally known, in an image forming apparatus such as a copying machine, which comprises an image reading unit and a print unit for performing image formation on the basis of data read by the image reading unit, as one basic method for faithfully reproducing a read original image in a print out operation, the image processing condition is calibrated. For example, N sets of input signals An having known density values are input as print data, and the print unit forms specific patterns on a recording medium on the basis of the input signals. The patterns on the recording medium are read by the reading unit as luminance data, and the read data are subjected to image processing such as masking processing to obtain signal values Cn as print data. The image processing condition such as masking coefficients are calibrated by, e.g., a method of least square so that the signal values An roughly equal the obtained signal values Cn. With the above arrangement, a read image can be faithfully reproduced, and desired image quality can be obtained in the print out operation.
However, in the conventional image forming apparatus, the reading state and the print state of the reading unit and the print unit may vary even slightly upon calibration of the image processing condition. In particular, the print unit relatively easily causes such variation. For this reason, even when the image processing condition is calibrated, the calibration result depends on, e.g., the state of the print unit at the time of the calibration. Therefore, when the state upon execution of image formation in practice is different from that upon calibration, appropriate image processing is disturbed.
The above-mentioned variation in the image forming apparatus often appears as local spatial errors. That is, the reading state and the print state of the reading unit and the print unit are not uniform, and the reading state may vary in units of portions of, e.g., the reading unit. In this case, the calibrated image processing condition may become improper for a given image forming portion.